Re: [PATCH 1/1] regulator: max77802: set opmode to normal if off is read from hw

[Tomasz Figa <tomasz.figa@gmail.com>]

On 27.08.2014 20:37, Mark Brown wrote:
> On Wed, Aug 27, 2014 at 08:32:38PM +0200, Tomasz Figa wrote:
>> On 26.08.2014 13:37, Javier Martinez Canillas wrote:
> 
>>> +		/*
>>> +		 * If the regulator is disabled and the system warm rebooted,
>>> +		 * the hardware reports OFF as the regulator operating mode.
>>> +		 * Default to operating mode NORMAL in that case.
>>> +		 */
>>> +		if (val == MAX77802_OPMODE_OFF)
>>> +			max77802->opmode[id] = MAX77802_OPMODE_NORMAL;
>>> +		else
>>> +			max77802->opmode[id] = val;
> 
>> I might be missing something, but I believe I see a flaw here. If after
>> a cold boot opmode is something else than OFF or NORMAL, the kernel
>> starts, turns the regulator off and does warm reboot, the setting is
>> lost and the regulator is switched to NORMAL mode.
> 
> That's essentially the situation the patch is trying to fix - if we boot
> and the regulator is off there's no way to figure out what the operating
> mode would have been so we have to pick something.  If you've got an
> idea for something better to do...
> 

Probably the only way to correctly handle this is to specify the right
operating mode in DT (after defining a binding for it).